Ghent University Academic Bibliography

Advanced

Dynamic Bayesian Networks for Acquisition Pattern Analysis: a financial-services cross-sell application

Anita Prinzie UGent and Dirk Van den Poel UGent (2009) LECTURE NOTES IN ARTIFICIAL INTELLIGENCE. 5433. p.123-133
abstract
Sequence analysis has been employed for the analysis of longitudinal consumer behavior with the aim to support marketing, decision making. One Of the Most popular applications involves Acquisition Patiern Analysis exploiting the existence of typical acquisition patterns to predict Customer's most likely next Purchase. Typically. these cross-sell models are restricted to the prediction of acquisitions for a limited [lumber of products or within product categories. After all, most authors represent the acquisition process by an extensional, unidimensional sequence taking values from a symbolic alphabet. This sequential information is then modeled by (hidden) Markov models Suffering from the state-space explosion problem. This paper advocates the use of intensional state representations exploiting structure and consequently allowing to model complex sequential phenomena like acquisition behavior. Dynamic Bayesian Networks (DBNs) represent the state of the environment (e.g. customer) by a set of variables and model the probabilistic dependencies of the variables within and between time steps. The advantalges of this intensional state space representation are demonstrated on a cross-sell application for a financial-services provider. The DBN models multidimensional Customer behavior as represented by acquisition, product ownership and covariate sequences. In addition to the ability to model structured multidimensional, potentially Coupled, sequences, the DBN exhibits adequate predictive performance to support the financial-services provider's cross-sell strategy.
Please use this url to cite or link to this publication:
author
organization
year
type
conference
publication status
published
subject
keyword
SEQUENTIAL INFORMATION, PRODUCTS, MODELS, MARKOV
in
LECTURE NOTES IN ARTIFICIAL INTELLIGENCE
Lect. Notes Comput. Sci.
editor
S Chawla, T Washio, SI Minato, S Tsumoto, T Onoda, S Yamada and A Inokuchi
volume
5433
issue title
NEW FRONTIERS IN APPLIED DATA MINING
pages
123 - 133
publisher
Springer
place of publication
Berlin, Germany
conference name
International Workshop on Algorithms for Large-Scale Information Processing in Knowledge Discovery
conference location
Osaka, Japan
conference start
2008-05-20
Web of Science type
Proceedings Paper
Web of Science id
000265665100011
ISSN
0302-9743
ISBN
978-3-642-00398-1
language
English
UGent publication?
yes
classification
P1
id
706903
handle
http://hdl.handle.net/1854/LU-706903
date created
2009-06-23 15:21:08
date last changed
2015-06-17 11:16:10
@inproceedings{706903,
  abstract     = {Sequence analysis has been employed for the analysis of longitudinal consumer behavior with the aim to support marketing, decision making. One Of the Most popular applications involves Acquisition Patiern Analysis exploiting the existence of typical acquisition patterns to predict Customer's most likely next Purchase. Typically. these cross-sell models are restricted to the prediction of acquisitions for a limited [lumber of products or within product categories. After all, most authors represent the acquisition process by an extensional, unidimensional sequence taking values from a symbolic alphabet. This sequential information is then modeled by (hidden) Markov models Suffering from the state-space explosion problem. This paper advocates the use of intensional state representations exploiting structure and consequently allowing to model complex sequential phenomena like acquisition behavior. Dynamic Bayesian Networks (DBNs) represent the state of the environment (e.g. customer) by a set of variables and model the probabilistic dependencies of the variables within and between time steps. The advantalges of this intensional state space representation are demonstrated on a cross-sell application for a financial-services provider. The DBN models multidimensional Customer behavior as represented by acquisition, product ownership and covariate sequences. In addition to the ability to model structured multidimensional, potentially Coupled, sequences, the DBN exhibits adequate predictive performance to support the financial-services provider's cross-sell strategy.},
  author       = {Prinzie, Anita and Van den Poel, Dirk},
  booktitle    = {LECTURE NOTES IN ARTIFICIAL INTELLIGENCE},
  editor       = {Chawla, S and Washio, T and Minato, SI and Tsumoto, S and Onoda, T and Yamada, S and Inokuchi, A},
  isbn         = {978-3-642-00398-1},
  issn         = {0302-9743},
  keyword      = {SEQUENTIAL INFORMATION,PRODUCTS,MODELS,MARKOV},
  language     = {eng},
  location     = {Osaka, Japan},
  pages        = {123--133},
  publisher    = {Springer},
  title        = {Dynamic Bayesian Networks for Acquisition Pattern Analysis: a financial-services cross-sell application},
  volume       = {5433},
  year         = {2009},
}

Chicago
Prinzie, Anita, and Dirk Van den Poel. 2009. “Dynamic Bayesian Networks for Acquisition Pattern Analysis: a Financial-services Cross-sell Application.” In Lecture Notes in Artificial Intelligence, ed. S Chawla, T Washio, SI Minato, S Tsumoto, T Onoda, S Yamada, and A Inokuchi, 5433:123–133. Berlin, Germany: Springer.
APA
Prinzie, A., & Van den Poel, D. (2009). Dynamic Bayesian Networks for Acquisition Pattern Analysis: a financial-services cross-sell application. In S. Chawla, T. Washio, S. Minato, S. Tsumoto, T. Onoda, S. Yamada, & A. Inokuchi (Eds.), LECTURE NOTES IN ARTIFICIAL INTELLIGENCE (Vol. 5433, pp. 123–133). Presented at the International Workshop on Algorithms for Large-Scale Information Processing in Knowledge Discovery, Berlin, Germany: Springer.
Vancouver
1.
Prinzie A, Van den Poel D. Dynamic Bayesian Networks for Acquisition Pattern Analysis: a financial-services cross-sell application. In: Chawla S, Washio T, Minato S, Tsumoto S, Onoda T, Yamada S, et al., editors. LECTURE NOTES IN ARTIFICIAL INTELLIGENCE. Berlin, Germany: Springer; 2009. p. 123–33.
MLA
Prinzie, Anita, and Dirk Van den Poel. “Dynamic Bayesian Networks for Acquisition Pattern Analysis: a Financial-services Cross-sell Application.” Lecture Notes in Artificial Intelligence. Ed. S Chawla et al. Vol. 5433. Berlin, Germany: Springer, 2009. 123–133. Print.